DragonFly BSD
DragonFly users List (threaded) for 2005-03
[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]

Re: Differences between AMD and Intel CPUs [was: Re: Dragonfly andHyperthreading....]

From: "Thomas Edward Spanjaard" <t.e.spanjaard@xxxxxxxxxxxxxxxxxx>
Date: Fri, 11 Mar 2005 19:05:07 +0100

<EM1897@xxxxxxx> wrote in message news:6C3B2051.2AA8F267.000424FF@xxxxxxxxxx
Well sorry, but you are just completely wrong about
everything. I've already cited the P8SCi as an example,
Im sure you know how to use google to get prices.

I was talking about both a CPU and the motherboard which you were claiming. In your P8SCi example, you left the CPU out of the price.

Your "test" is also completely wrong, and your analysis
of the bus wrong as well. For example, passing 80Mb/s
through a box with a 33Mhz/32bit nic might use 50% of
the cpu, and you pop a pci-x card of the same type (say
the em driver) and it will drop to 1/3rd of the usage.
Saying that you can pass some amount of data is not a
useful analysis, because if you use 100% of the cpu
instead of 30% to achieve the throughput its worthy
of note. There are 1000s of I/O operations that are not happening in the background, and as the bus saturation increases,contention increases and those operations take longerand longer. Its not rocket science, its very easy to
look at the interrupt load with different kinds of
cards. Anyone that thinks the way you do just isn't
paying attention. But, its very typical of what I've
been saying. That is, that a lot of people who should
know better don't seem to understand the basics of
what the real bottlenecks are. It seems almost comical
that teams of engineers are rewriting an OS trying to
squeeze some cycles out of a CPU without any understanding
of the main bottleneck to networking performance.

That's not a difference between PCI-X and PCI, but a difference in adapters and drivers. Lookup the specs, PCI-X doesn't provide the functionality you claim. To be able to test both buses properly, you need to run an identical adapter in it (PCI-X happens to be backwards compatible with PCI2.x, so we're lucky there). Only then you can bench both properly, as they'll be using the same chip and driver.

Furthermore, I was just pointing out that you can achieve good results with PCI, and never cared about CPU utilization. In fact, it was on a UP PIII-S 1.2GHz system, and I got nowhere near 50% CPU utilization from the I/O. See my remark about properly using the hardware's abilities.

Also, why do you ignore my remark about your client software?

-- Thomas E. Spanjaard

[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]